Disease Prediction based on Retinal Images

The paper's major goal is to create a basic framework for recognising disorders including diabetes, hypertension, and heart attacks. In the retinal fundus image shows two types of blood vessels that is arteries and veins. To diagnosis the different types of diseases, it is more important to distinguish the vessels into arteries & veins, which are highly correlated with brain and heart functioning. Preprocessing steps are introduced for the segmentation vessel using median filter algorithm. After pre-processing the images are undergone with Convolution Neural Network (CNN) approach. CNN is used to recognize the key feature of the images and they are differentiated based on the key feature. When we give the input of retinal images the difference in the retinal eye was observed based on their diseases. There are numerous diseases like heart attack, diabetics etc, which kills the people. With the help of this project some of the diseases are predicted with the help of the retinal eye. Retinal eye is very important which consists of retinal nerves which are connected to the brain. The VGG 16 architecture is used to find the diseases using retinal images.